From the answers you choose to the result you read

87 questions in 3 stages

The first stage asks 50 questions about five personality traits: sociability S/R, emotional response L/C, organization O/U, cooperation A/E and curiosity I/N. Of these, 20 are anchor questions that ask directly about a single trait. The other 30 describe everyday scenes and ask what you would do.

The next 25 questions look at what moves you. There are 15 questions about nine core desires, 7 about where you put your energy first through self-preservation, social and one-to-one instincts, and 3 to distinguish two desires that are easy to confuse. Similar actions can come from different priorities, so this stage looks beyond the action itself.

The final 12 questions explore how your reactions change with the situation. You see a summary after the first stage, and each later stage adds another layer to the reading. A free account is needed to read the full interpretation.

Social and private you

Every question is tagged either ‘around other people’ or ‘on your own’. We calculate scores separately for these contexts, producing a five-letter code for your social self and another for your private self. We keep the answers apart rather than averaging away a possible difference.

Different codes do not mean you answered incorrectly. The difference is information about how you respond in each setting. The question draw balances both contexts and the five traits, so you can read the two views side by side without treating either as the more authentic you.

Questions and options

The first-stage bank contains 236 questions: 98 anchors and 138 everyday scenes. We draw 50, balancing the five traits and both contexts. The next stage draws from 137 questions, made up of 87 desire questions, 35 instinct questions and 15 separators; new tests include at least one dedicated question for each of the nine desires.

A random key created when you start determines your questions. The same key always produces the same questions, so pausing and returning does not change them. Answer options are shuffled within each question, and the score uses the content you select rather than its screen position.

When the bank changes, the updated questions apply only to tests started afterwards. Your key records the bank version, so a test in progress finishes with the questions it started with. If you retake on another version, some of the difference in your result may come from the changed questions.

What scores mean

Each option carries weights for the traits, and we add your chosen weights within each context. Choosing the average weight of each question's options would give 50; always choosing the strongest option in one direction would give 100 or 0. Values in between are scaled separately on each side of the midpoint.

These scores are not percentiles: they do not say you score higher than a percentage of other people. We do not yet have enough comparison data for that. Distance from 50 describes how consistently your answers leaned in one direction, rather than where you rank among other test takers.

Traits scoring from 45 to 55 are marked borderline. In that range, one or two different answers can change the corresponding letter in your code. Read the score alongside the letter, because a change in label can hide how close both results are to the middle.

The self-gap index

The self-gap index runs from 0 to 100 and combines differences between the two contexts. Trait-score differences have a weight of 0.5, instinct differences 0.2 and situational-reaction differences 0.3. It is more than a count of different letters in the two codes.

For each trait, the first 10 points of difference are treated as measurement wobble and do not count toward the gap. A 50-point difference fills that trait's share of the index. This is a scoring rule for handling small differences, not a claim that they cannot matter in your life.

Neither a high nor a low index is better. A higher value means your answers showed more difference across contexts; a lower value means they were more alike. It does not assess your authenticity or mental health, so use the accompanying readings to explore what the difference means to you.

Your leading desire

The nine desires do not appear equally often across questions. To avoid favouring a desire simply because it appeared more, we divide its score by how much that desire could vary on the questions actually drawn for your test. The highest adjusted score becomes the main desire.

If the leading scores are close, we mark the result as low confidence. Two easily confused desires also receive dedicated questions to help distinguish them. The leading label is a starting point for reflection, with room to consider a nearby desire too.

Comparing reactions

Situational reactions are summarised as one of four patterns: much the same everywhere, different with familiar people and strangers, different when demands rise, or switching modes between settings. Rising demands include being tired, rushed or facing something hard. The pattern describes the clearest change in your answers, rather than a better or worse way to respond.

New tests pair similar scenes across the social and private contexts, with options built in the same way. For example, one scene asks about a friend's long phone call on a worn-out day, while its partner asks about chores you have put off on a worn-out day. Pairing is meant to measure both settings with the same ruler rather than mix differences in question design with differences in your reactions.

Same-position choices

In the full test, a note appears above your result if you answered 60 or more questions and at least 45% of your choices occupied the same screen position; for 40 to 59 answers, the threshold is 50%. For the 10-question taster, it is 9 of 10 choices in one position. The note includes a way to retake, but it does not change your scores or decide whether your answers are valid. That judgment remains yours.

Timing and simulations

On 2026-10-03, we measured 56 real completed stages, excluding automated runs. Median times were about 16 minutes for the first stage, 8 minutes for the next and 3 minutes for the last, with roughly 19 seconds per question. The progress bar uses this pace for its time hint; it is an estimate from observed answers, not a deadline.

We ran thousands of simulated respondents through the real questions and scoring code, including some who answered as the same person in both contexts and some as two different people. With the current bank, when one of each is paired at random, the ‘two different selves’ respondent gets the higher gap index about 98% of the time. The situational component alone separates them about 83% of the time.

Limits of these checks

Simulated respondents choose using the same weights that the scorer uses, so the check demonstrates internal consistency only, not accuracy for real people. Drawing on Big Five, SLOAN, Enneagram and Mischel and Shoda's CAPS does not establish that this test is scientifically validated. We do not yet have enough data on real response distributions, stability on retaking or agreement with other tests.

This is a self-reflection tool, not a diagnosis or an assessment for making decisions. If a reading feels unlike you, use the ‘fits me / a bit different’ buttons below each reading to tell us. We use that feedback to improve questions and wording, and you do not need to reshape your view of yourself to fit the result.

Research we draw on

The five traits follow the SLOAN model from Big Five research, and the nine desires and three instincts follow the Enneagram. Situational reactions draw on Mischel and Shoda's cognitive-affective personality system, CAPS (1995). Measuring the social and private contexts separately is this service's own design built on that research, rather than an independent academic theory.

The birth-date reading offered elsewhere on this site is separate content calculated from birth details. It plays no part in these personality scores, which come from your chosen answers. If you read both, keep their sources distinct: a birth-date reading is not evidence for a personality-test result.